gussetplate: how to weld

[lolobau]

Good Day

I would like to ask something about the welding of gussetplates

please see attached photo

does a gussetplate needs to be weld from both sides or only from one side?
is it correct to interrupt the welding joint for 80mm sections? this means if the gussetplate is for exaplae 500mm lpong that i have only a welding section of 80mm in front and 80mm in the back

regards

lolobau

 

[lolobau]

here one other photo

 

[lolobau]

here one other photo of an sample

http://files.engineering.com/getfil...e-4cb0-ad03-cb0ec5d45dd2&file=PC040055bbb.jpg

 

[JAE]

Typically gusset plates are welded on both sides to create a concentric/centered load application on the attached elements.

In your photos - this does not look like a very significant structural assembly so I would probably not worry about it too much.

 

[lolobau]

hhmm... it makes sense ja that the loads will be distributet better and symetric from bnoth sides. this is a small truss only 1400mm but i will have also some which are 7000mm so i want it to be totaly correct so i woudl appreciate all the good imput how to make a propper gussetplate

 

[TXStructural]

Gusset plates usually are intended to impart flexibility to the connection. What you have is a fully fixed end condition for each member. The way the gusset is attached to the vertical and horizontal members, the tension on the diagonal will pull between the nearest points of attachment - think of the gusset as elongating under tension, with more load moving between closer welds, and less between more distant welds. As the diagonal pulls and tries to deform the truss shape, the gusset in your photos may try to buckle on one side of the diagonal and stretch on the other, imparting moment into the vertical and horizontal members at the location of the welds.

 

[lolobau]

to TXstructural: yes you are right all this shoudl be also considdered. I mean such an roof truss is actually everytime calculated and designed as a ONLY AXCIAL stresses model but usually everytime the contractors provide gusset plates, istn´t it?

 

[lolobau]

are there no nice photos or samples of a gussetplate online?

 

[BAretired]

Ordinarily, a truss is symmetrical and the gusset plate is welded to members on each side. In your case, the truss is unsymmetrical so the members are loaded eccentrically whether you weld on one side or both sides. Welding on the back side produces more eccentricity on each member and is not as good as welding on the front. Welding across members as shown by the two yellow arrows is not considered good practice because it weakens the member.

The gusset plate in the photo is not shaped to permit a balanced weld on the diagonal. It should be shaped to permit more weld on the heel and a lesser amount on the toe.

BA

 

[Ron]

Gusset plates usually are intended to impart flexibility to the connection.

???? In truss applications, they do nothing for the flexibility of the connection since the connections are considered to be in tension or compression. In column applications, they increase the strength of the connection to resist bending...therefore making the connection more rigid and transferring stresses away from the basic column to baseplate weld. Gusset plates reduce the flexibility of such connections.

 

[lolobau]

thanks for the new tipps

I would like to ask something BA
just to make sure

what do you mean with "gusset is not shaped to permit a balanced weld"


I attach a new photo with some differend cases A, B, C, D and differend arrow colurs so that we can discuss it better

 

[lolobau]

in reference to the attached photo

"A“ red, violet, green are cross wildings and weaken the section? But yellow and blue is correct?

“B” there should be actually a welding in red and violet?

“C” blue and violet is perfect but yellow and red wrong?

“D” yellow, green white should be rather provided and blue and red left out?

 

[BAretired]

In the usual case, a gusset plate connects a pair of angles, one on each side of the plate. The "connected" leg of each angle is in the same plane as the gusset. The "outstanding" leg is normal to the gusset and is not connected at its free end.

Welds are "balanced" when their centroid is aligned with the centroid of the angle. They cannot be completely balanced because one toe is not welded, but they can be nearly balanced. Suppose you have calculated that you need a total length of weld of 200mm for an equal leg angle. A completely balanced weld would be 50mm at each toe and 100mm at the heel. Because one toe cannot be welded, a nearly balanced weld would be 150mm at the heel and 50mm at the connected toe.

In the case of an unsymmetrical truss (not something I normally specify) you must attempt to place the weld metal in a similar fashion. I will use another post to comment on your photos.

BA

 

[BAretired]

Others may have a different opinion, but for an unsymmetrical truss, here is what I would prefer:

Photos B and C

1. Move gusset plate to the other side of the connected leg of the top chord and end vertical, terminating it at the dashed chalk lines shown on Photo B.

2. Do not trim leg of diagonal. Terminate angle in a 90 degree cut.

3. Reduce height and increase length of gusset plate as required to accommodate a balanced weld for diagonal. The weld along the chalk line is already balanced so no need to do anything more.

4. Yellow arrow Photo B...butt weld is okay but probably not necessary. Blue arrow Photo B...delete. Red arrow...use 1/4 L_w fillet weld. On heel, use 3/4 L_w (no arrow shown).

5. On Photo C, violet and blue arrows are okay but will be on opposite side of angle leg. Yellow and red arrows...do not weld here as it weakens the member.

Comments would be similar for Photos A and D.

BA

 

[lolobau]

To BAretired:

thank you really realyl a lot for your help. Now you opend my eyes for all the details.
To be honest I did never considder it with the balanced weld on the hell and the toe
I see on the photo "B" how you distribute it with the wleding lenght of 1/4 outside and 3/4 inside the angle.
Make sense somehow but do the guys really do it this way on the site?
Does it have a negative influence if the outside welding which was supposed to be 1/4 will be 4/4?

see attached graphics. Is this now correct?

http://files.engineering.com/getfile.aspx?folder=717d11b5-e62a-450c-9910-5cc16b0c9015&file=FF2.jpg

http://files.engineering.com/getfile.aspx?folder=de5f6d7b-973e-4d8c-8bad-f034cad271fc&file=GG2.jpg

 

[BAretired]

Attaching the angle to one toe is poor practice and puts too much eccentricity on the angle. That would be a negative effect in my view.

Sketch FF2 with gusset outside looks okay as drawn.

Sketch GG2 with gusset inside indicates the gusset plate running to the face of the outstanding leg of the angle on the top and right. This won't work because the plate will run into the curved angle fillet. Keep the gusset and weld clear of the fillet. If the angle is large enough, terminate the gusset plate at the c.g. of the angle and use L_w on that line, otherwise move the gusset to the outside and follow Sketch FF2.

BA

 

[TLHS]

Honestly, unless the width of the truss is a huge concern I likely would have flipped the diagonal to the backside of the gusset so it gets out of the difficult geometry of that joint. If the brace came in at 45 degrees, it would be fine to put it on the same side, but from the pictures it looks like it's more like 20 degrees, which starts making the geometry ridiculous ridiculous. Then where 90+- degree corners happen, I probably would have just stopped the angles clear of each other and passed all the load through the plate.

You get a little bit more inter-member eccentricity to deal doing this, but your fab is much easier.

I wouldn't do this with something heavily loaded, because the plates would start getting ridiculous.

I know you're now working with sketches where you're not welding the right angled members directly together. I'll just reinforce that by pointing out that you're not really trying to pass forces between those two members. They're at close to right angles and don't interact significantly. If there's no significant force transfer there, why bother with a connection that's a significant fraction of the capacity of the member.

 

[lolobau]

To BARetired: So FF2 is correct? Also with the ¼ and ¾ LW? But would it be bad if both welding would have the same length? Or is it just to save the welding afford?

To TLHS: yes this is what I am trying to show in GG2 that the diagonal member is placed, mirrored to the other side of the gusset. The angel is approx 20 degree you are right and this makes the gusset plate very long.


lolobau

 

[BAretired]

Both welds can be the same length on FF2, namely 3/4 LW.

BA